Kimmono (@kimmono) profile picture

Support @Arsenal, interests: UAP, AI, Ukraine and Energy.

Oslo, Norway
Joined February 2008
Photo by Kimmono (@kimmono)Photo by Kimmono (@kimmono)Photo by Kimmono (@kimmono)Photo by Kimmono (@kimmono)Photo by Kimmono (@kimmono)Photo by Kimmono (@kimmono)Photo by Kimmono (@kimmono)Photo by Kimmono (@kimmono)Photo by Kimmono (@kimmono)Photo by Kimmono (@kimmono)
The @EU_Commission should also investigate the 10x price increase on DDR5 memory! This is a disaster for innovation and privacy in Europe, as it is too expensive to buy hardware for Local AI, PC and servers. @vonderleyen
A federal court in California found that the plaintiffs’ allegations plausibly suggest Samsung, SK hynix, and Micron agreed to restrict DRAM supply. The court denied the three companies’ motion to dismiss, allowing the antitrust lawsuit to proceed.
1
62
Six of Nvidia’s seven biggest customers are building their own chips. “NVIDIA’s seven biggest customers account for the vast majority of GPUs being deployed: Google, Microsoft, Meta, Amazon, SpaceX, Anthropic and OpenAI. All of them, except SpaceX, are making their own chips.” “If NVIDIA didn’t prop up CoreWeave and the other neoclouds, those industries just would not exist, and the hyperscalers would have full control.” “Jensen’s the only executive in the semiconductor industry convincing supply chain companies to invest like they’ve never done before. From our benchmarking, the TPU is just as good as NVIDIA. But Google doesn’t have that DNA of wheeling and dealing.”
137
248
64
2,468
338,739
Anthropic reported a similar sudden and unusual jump in robotics capabilities with Fable 5.2. As the general intelligence of frontier models scales up, it is increasingly surpassing specialized training. This will ultimately generalize to every sphere of human ability.
Fortune is reporting that OpenAI is so convinced that LLMs will become brains for robots that it has disbanded its robotics team.
87
189
40
2,189
130,279
Yesterday, before digging into The Flight Situation, I wrote: "If the RADARs were being jammed, that should be detectable, that wouldn't be a mystery. Same with broad-spectrum comms jamming. So, if it's not from an exterior source, that points to widespread system degradation; and I can't *imagine* a way that happens accidentally (unless an implementation is mind-bendingly stupid, in a way I genuinely struggle to fathom)." Hoo, boy. I was hoping for more info today but, IMO, the important info isn't from today at all. We have a full on Story Time on our hands. But, first, the immediate updates: Good news is that the FAA's comms are is back online, flights are moving again. The bad news is that "mind-bendingly stupid" was... kind of an understatement. I was not pessimistic enough, and that's truly saying something. Would it be embarrassing if this was the result of ignoring 20 weeks of warnings? This one's gonna get a *little* less objective as we go. Starting with the objective stuff: 1. What's up with flights? The nationwide ground stop ended Friday at 8:15 PM Eastern, after four hours and some change. Unfortunately, that doesn't mean the airlines are back to normal. More than 9000 commercial flights (and 3000 private ones) were canceled yesterday, and another 4–5000 are being canceled today, with who-knows-how-many-thousands more delayed. The problem now is mostly logistical. Aircraft and crews are scattered across the country, passengers have missed connections, and airlines have to rebuild their schedules in light of the, uh, shitshow. They can't just resume every canceled flight: the plane and crew that were supposed to run it may just be on the other side of the country (they were stranded, yesterday, too). And, unlike most industries, "just fly overtime" isn't on the table. Pilots and flight attendants operate under strict FAA duty-time and mandatory rest rules. Once a crew "times out" waiting through a four-hour ground stop, they're legally grounded until they get mandatory rest — and with airlines already running on tight pilot staffing (guess those chickens are coming home to roost), there just isn't a deep bench of spare crews to swap in. But, all that said, the FAA expects schedules to return to normal by the end of Sunday (so, I expect Tuesday), although some unlucky winners are probably going to see this run longer. Sorry. If you're planning on flying this weekend, check your flight status directly with the airline before heading to the airport (I would like to say "surely I don't need to write this", but). 2. So, what broke? We now have a (hilarious, watch the video) preliminary explanation from Transportation Secretary Sean Duffy and the FAA. Yesterday, a contractor called L3Harris was performing routine maintenance at a remote FAA facility, migrating telecommunications connections from 1980s-era copper lines to fiber. This did not go as planned. A new router started flooding the FAA's private network with "over 1000 connection attempts every few seconds" (either because of installer error or faulty hardware, they're still figuring it out -- but, looks like not a cyberattack! 🎊). Now, if you know networking, 1000 connection requests every few seconds isn't a lot. In 2026. A cheap cloud server shrugs that off. Your phone can do that. But, because it was interfacing with hardware from the dinosaur era — aging routers and switches built for tiny, predictable aviation circuits — *now* "1000" is a lot. The modern hardware overwhelmed the ancient hardware, causing widespread instability in the communications and radar data reaching air traffic controllers. Importantly, this does NOT mean all the physical radar installations stopped working. The sensors were still collecting information; the network just wasn't delivering that information to the people that needed it. That's consistent with what we found yesterday: the external radar, GPS, and weather data I checked looked normal, even as the national air traffic system was shutting down. The new equipment reportedly remained operational; the old infra just couldn't deal. Once engineers isolated the affected work, the network stabilized. Sick. 3. What don't we know? The FAA hasn't completed its root-cause investigation (the FAA is thorough: this will probably take time). Officials are still examining whether this was contractor error, faulty equipment, or some combination of the two. "No cyberattack" is holding up, god bless. But, I think there's a much more important question than which router malfunctioned: How in, pardon, the *actual* fuck did maintenance at one (1) remote facility cause a nationwide (🇺🇸) failure? I used to be an analyst; I'm currently a software dev. We *have* SOPs for isolation, unit tests, failovers, and rollbacks *because* we know things go wrong. This is like Engineering 101. Shit happens, bring a raincoat. Equipment fails. People get hungry or sleepy or pissy and make mistakes. Sometimes, there's just some genuinely unexplainable bullshit: god has a sense of humor, and he's laughing at developers. That's the entire reason critical infra is supposed to have fault isolation, redundancy, tested fallback paths, and rollback procedures. We don't yet know whether those safeguards were absent yesterday, if they failed, or if someone just didn't use them correctly. We *do* know that whatever safeguards were in place, if any, didn't *work*. Sometimes, it's some genuinely unexplainable bullshit: a real "who could've seen this coming?" This was *not* one of those times, lmao. 4. How did we get here? On October 31, 2005, the FAA suffered an outage during a telecommunications migration at Chicago O'Hare: some comms gear failed, controllers lost radar picture (equipment still working, they just couldn't see it: sound familiar?), and air traffic had to be restricted. The Department of Transportation's Inspector General investigated and published its findings in April 2006 (Report AV-2006-047). Among the failures it identified: inadequate coordination with local engineers, incorrect equipment requirements, and a... "problem with the backup system" isn't quite right. That sort of implies the backup system exists. When the FAA tried to switch to its backup, it discovered that the backup service wasn't available because: nobody had tested and verified that the backup line actually existed. So, setting aside objectivity for a second: Are you fucking kidding me? lmao? The IG (inspector general) recommended "improving transition planning, verifying site requirements, and ensuring that migrations didn't disrupt air traffic operations". Would it be embarrassing if the warnings were ignored for 20 weeks? How about 20 *years*? 5. Oh it gets worse. 2023: an independent FAA safety review warned that the national telecommunications backbone was "dangerously obsolete". The report explicitly identified budget constraints as a reason the FAA remained dependent on ancient technology. It said the FAA had *already* classified the infrastructure risk at its highest level. We're not done: in September OF THIS YEAR, the Government Accountability Office (GAO) published another warning (GAO-26-107992) about the accelerated $12.5 billion modernization program: among other things, they found that the FAA lacked a reliable overall cost estimate, and had 11,000+ individual project schedules that hadn't been integrated into a master schedule. GAO specifically warned that this made it harder to "coordinate installations" and "minimize disruption to air traffic operations". So: who could've seen this coming? Hellen fucking Keller could've seen this coming. None of this establishes *which* management issues contributed to yesterday's... situation... but, the fact that there are "too many issues to narrow it down to one" is actually a joke. 6. Now what? I think two things can be true simultaneously: A: The FAA desperately needs to replace this aging infrastructure. Decades of funding problems, procurement difficulties, and delayed modernization have created a genuinely dangerous situation. B: Replacing that infrastructure without adequate safeguards (read: yesterday) can *also* be a problem. Damned if you do, damned if you don't. The FAA, and aviation culture in general, are usually commendably safety-conscious (borderline neurotic, in a good way) — just like with crew rest rules, they don't fuck around. Their safety mechanism here was basically "this is sketchy, ground everything": working as intended, even if it never should have been required. Apparently, the same simply cannot be said of the infrastructure surrounding aviation. At this point, the FAA's investigation needs to establish exactly how this failure propagated, what redundancy existed (if any), whether fallback and rollback procedures were tested, and, ultimately, explain how a local maintenance operation could compromise operations nationwide. Yesterday, I said I couldn't imagine a widespread accidental failure unless the implementation was "mind-bendingly stupid". My mind is now bent. And, to be clear, the issue isn't "a technician made one stupid mistake". Shit happens. The part that really makes you wonder is how, despite twenty years of documented warnings, the system *still* didn't have a raincoat.
Sorry for the long-ass post, I did some OSINT shit. I pulled live public flight feeds, weather radar status, and space weather data to see what's actually going on during today's nationwide FAA grounding. TL;DR: radar, GPS, and weather sensors are all working fine. What broke is the long-distance communications network that lets the FAA's 22 regional air traffic control centers talk to high-altitude planes and hand them off to each other. When people think of air traffic control, they picture the glass tower at an airport. Those towers only handle takeoffs, landings, and nearby airspace. Once a plane climbs out, it's handed off to one of 22 regional Air Route Traffic Control Centers (ARTCCs) — giant facilities that manage high-altitude cruising traffic across whole regions of the country. Normally, those 22 centers are regionally isolated: if one loses power or a local fiber line, neighboring centers absorb or route traffic around it. Having all 22 degrade at once is virtually unheard of because it means a shared national backbone failed, leaving no healthy neighbor to take over. That's why the FAA's Command Center used such stark, blanket wording ("Due to extraordinary circumstances and for reasons of safety... all airports are ground stopped until further notice"*) — it's standard protocol when a national-level freeze has to happen immediately before engineers even know how long repairs will take ("safety issue" means "oh shit, stop" in aviation circles, unlike with AI companies). All 22 centers losing comms at once is, indeed, "extraordinary circumstances". Here's what the live data shows: 1. US airspace is emptying out fast. Normally on a Friday evening, thousands of planes are in the air over the US. In live tracking data from OpenSky Network, the number of airborne planes over the lower 48 is steadily draining as flights land and no new commercial departures take off: • 6:36 PM EST: 1,083 airborne • 6:42 PM EST: 1,035 airborne • 6:49 PM EST: 990 airborne There are zero emergency distress codes broadcasting. 2. Major airports are, predictably, stacked. Looking within 30 miles of major hubs around 6:37 PM EST (`adsb.lol`), almost nothing is moving: • NYC/JFK area: 8 in the air vs. 67 on the ground • LAX: 14 in the air vs. 40 on the ground • Dallas (DFW): 7 vs. 36 • Denver (DEN): 13 vs. 29 • Atlanta (ATL): 5 vs. 25 • Chicago (ORD): 9 vs. 24 • DC (DCA): 8 vs. 22 A lot of the coastal airports had to stop accepting diverts. 3. The few planes still taking off are enlightening. Across the entire US at 6:36 PM, only aircraft were taking off (that is: below 10,000 feet and climbing). They fell into two clear buckets: • 3 international widebody jets (`DAL281` out of Seattle, `DAL156` out of NYC, and `AFR71E` out of Newark), matching the FAA's note that international flights are being handled case-by-case. • Small private planes and a couple of business jets flying under VFR (visual flight rules) — essentially, staying below 18,000 feet, where pilots don't need en-route air traffic control clearances. Honorable mention to the 1 Coast Guard/military helicopter climbing near Tampa, out of 29 military aircraft tracked nationwide. 4. Weather radar isn't down, and there's no solar storm or GPS jamming. A lot of people wondered if today's FAA ground stop, rumors of a National Weather Service outage, and Microsoft's 365 outage were all connected. Checking the raw feeds: • Weather radar: 199 out of 208 US Doppler radar stations are actively sending fresh scans (within the last 15 minutes). The "NWS outage" rumors were website and caching glitches, not offline radar hardware. • GPS & Space Weather: NOAA space weather is completely quiet (`Kp = 0`), and aircraft GPS accuracy readings across the Midwest and Northeast look totally normal. • Local airport radar & towers: Local FAA ground-station radar broadcasts and local airport control towers are still up and transmitting. So, local sensors and local towers are fine. The failure is specifically in the long-haul telecommunications backbone connecting the FAA's 22 en-route centers to their remote radio sites — which is why en-route controller frequencies went quiet while local towers, weather radars, and low-altitude visual flights kept running. Federal officials say there's no sign of a cyberattack, so we're waiting on the FAA's technical post-mortem for what tripped the national network. That doesn't rule out a cyberattack as the root cause of the comms outage; it's a "we don't know". Shoutout to adsb.lol for their free API.
4
22
5
75
8,713
GROUNDSTOP. Let's talk about the FAA funding question. Sec. Duffy says he wants $50 billion and has pointed to a lack of Congressional funding as the issue. Here is the reality... thread.
20
144
11
567
83,789
Kimmono retweeted
After Democrats flipped both chambers of Congress in the midterm elections, the House Oversight Committee is considering a further investigation into the UAP files.
15
32
13
130
38,200
This is clearly not drones, this i UAP(s).
Footage of suspected drones near RAF Marham. Which is not far from Lakenheath/Mildenhall/Feltwell
1
98
News: The US intel community is still working to understand the relationship between a Chinese businesswoman who had an affair with Nvidia CEO Jensen Huang and the Chinese government, with multiple sources telling CNN that the intelligence collected on her to date does not prove that she was acting as a spy for Beijing. cnn.com/2026/11/12/politics/…
406
955
389
5,297
2,614,995
Kimmono retweeted
Nvidia chips are still scarce, even now. But to maximize that premium, Nvidia has to support the neoclouds. If the new Nvidia CEO turns even slightly skeptical about supporting the neoclouds, spot prices will take a hit. That will hurt neocloud margins. To avoid this, the neoclouds will have no choice but to reach out elsewhere. AMD, or the TPU camp. In other words, the neoclouds break free of Nvidia's grip, and Nvidia's dominance and influence both take a hit. That in turn means a hit to the massive order volume that underpins Nvidia's only moat, its control of the supply chain, and it means Nvidia's flywheel breaks. Jensen worked tirelessly to protect this flywheel. He pushed on many fronts to get a blind eye turned to neoclouds leasing capacity to Chinese hyperscalers, and he encouraged chip exports to China so that China would never forget the taste of the Nvidia sauce, working relentlessly to have China's AI ecosystem built on top of Nvidia. That way, even if the latest GPUs become hard to get inside China, there is still a reason to rent the same chips from overseas neoclouds. Keeping Chinese developers inside the Nvidia ecosystem was also a way of protecting neocloud utilization. Still, all of this was playing with fire along the edges of the gray zone, something deeply improper and dangerous in the eyes of Washington. Even so, Jensen did what was best for Nvidia. That cannot be denied. Jensen meant to wait until the neoclouds could order the next generation of GPUs on their own. He probably saw sharing risk with customers until then as being in Nvidia's interest too. What I want to know now is whether the next CEO can wait that long. According to the WSJ article... the current CEO does not seem to be that patient. I worry about an Nvidia without Jensen.
83
176
29
1,605
274,062
25 million Trump voters stayed home in the midterms, and 3.6 million switched sides.
29
373
30
884
24,429
As I predicted, Claude is off the leash. Fable 5.2 has solved two open problems in theoretical physics, one of them completely autonomously.
When you look at a frozen river, you must also imagine the current raging beneath the ice. I think all three big labs have achieved autonomous research and their internal models can now choose their own research directions. We will begin to see the fruits of this work this month.
76
233
27
2,143
88,215
Benchmarks:
Introducing Claude Fable 5.2 and Claude Mythos 5.2, designed to tackle the most daunting technical challenges. They’re capable collaborators on ambitious projects in many fields, and show improvements in areas like coding, reasoning, and vision.
17
24
4
424
33,714
Kimmono retweeted
President Trump says Air Force pilots told him in the Oval Office that a mysterious round object flew past an F-22 fighter jet at three times its speed, describing an encounter that defies conventional aerodynamics. Trump called the pilots "the real deal" and said they had shared "things that are very strange." He described himself as "not too much of a believer" but said the accounts from these aviators stood out. The object, he said, was round, not aerodynamic, and moved far faster than what he called "the best plane that anyone's ever seen." When asked whether a secret government program studying UFOs could exist without a president's knowledge, Trump said he hadn't heard of one, but pledged to find out. Full story here foxnews.com/us/trump-air-for…
81
280
22
1,606
208,742
Kimmono retweeted
China’s Space Program: Major Future Missions Revealed. 🧐 @NASAAdmin Source: teapawt.net/AJ_FI/status/211975945…
🤖 Made with AI
Slide on Chinese space timelines: Dec. 2027: Asteroid redirect mission 2028: Tianwen-3 Mars sample return, return 2031 2030: Tianwen-4 Jupiter mission By 2049: reach 80-100 AU, build a full-domain deep space exploration system. 2026-2034: large scale megaconstellation deployment
6
42
3,598
Across our projects @GoogleDeepMind, we are witnessing a genuine phase transition in the pace of scientific discovery. You will see a glimpse of this in our releases in the next few weeks.
301
784
285
9,372
1,184,928
Kimmono retweeted
David Glosser, Stephen Miller's uncle, told CNN his nephew is an "immigration hypocrite": Our family history is entirely dependent on our ability to escape a terrible situation in Belarus. Had we not been able to enter America when we did, Stephen Miller would not exist. Everybody else who didn't make it to America ended up dying in the Holocaust.
286
5,938
283
17,012
355,023
Kimmono retweeted
Yesterday I posted about working with an AI scientist to find new ways to fight my cancer. A chemistry professor at the University of Michigan saw it and emailed me. Matt Soellner @mbsoellner works on the exact kind of lung cancer I have. He offered to synthesize compounds and test them in cells and animals. A lightbulb 💡 goes off above my head. Obviously...I MUST introduce Matt to Jan. Andddd by this morning, they're talking about how to take an AI-designed molecule off a computer screen, make it in a lab, and find out whether it can fight my cancer. EXCUSE ME!? Less than 24 hours after my post. We haven't made a drug. We don't know whether any of this will work. Making and testing compounds will take months. But Matt saw a patient trying to find a way forward and chose to help. He brought the ability to make and test things that might otherwise never leave a computer screen. I'm sharing this as it happens because this is what gets me excited about AI. I can put my story out there, connect people with different skills, and help move an idea toward an actual experiment against cancer. I deal with the every day bullshit from cancer as a patient but have realized there's an army of incredible people out there working hard behind the scenes on treatments and cures for people like me and millions of others. It's an honor and a privilege to bring their efforts out for the world to see. AI is accelerating what these scientists can do. I'm just the human meat proxy making introductions and sharing the story. But if that helps get a promising idea off a screen and into a lab faster, I'll take it! 🥩
So get this...a stranger DM'd me on X and offered to help design my backup cancer drug. For free! His name is Jan Domanski @jandom_random. He runs science at @open_fold, an open-source AI model that predicts how proteins and drugs physically fit together. My cancer is driven by one broken protein called ALK. Lorlatinib (the daily pill I take) jams itself into it and shuts it down. It's working. But resistance isn't an if, it's a when. The tumor mutates the pocket and the drug stops fitting. So Jan started modeling it on his own time, burning his own tokens. He ran AI models against ALK and lorlatinib, mapping where the known escape mutations land. And what did he find? Most of them hit right in the drug's pocket, swapping small pieces for bigger ones that physically block the drug. That's the good kind of bad news. A pocket problem is something you can design around. But the normal route to a new drug is ~2 years and ~$4-5M just to be ready for human testing. Jan's dream is getting a medicine to one patient in ~1 month. I'm not taking a homemade molecule while lorlatinib works. But if it ever fails....shit...I've got a guy! I started posting actively on X just a few weeks ago and it has brought me together with the most amazing, generous people!
32
56
13
541
31,755
Thank you, Mr. Kushner. Those five principles are important. However, Moscow subscribed to these principles in documents signed in 1975 (Helsinki Final Act) and 1990 (Charter of Paris). The problem isn't agreeing to principles, but Moscow's failure to abide by them. 🧵 /1
President Trump today advanced an important discussion with several European leaders about the need for a broader European security architecture that can help establish the conditions for lasting peace. Building on his leadership and the substantive work undertaken this week in Brussels with Ukraine, NATO, and our European partners, @SEPeaceMissions and I wanted to share some of the objectives guiding our ongoing discussions. For more than 30 years, Europe and Russia have struggled to establish a sustainable security framework. Decades of mistrust, competing security concerns, and the devastating consequences of this war make this an extraordinarily difficult challenge, but one that is essential to lasting peace. Our objective is to develop practical arrangements that address the security concerns of all sides, reduce the risk of future conflict, and create a more stable and predictable security environment. We are exploring five principal goals: permanent security consultations, reliable crisis communications, greater military transparency, reciprocal measures to reduce escalation risks, and an enduring framework for long-term European security and stability. Any arrangements must fully respect the sovereignty, territorial integrity, and independent decision-making of every nation. Each country will retain authority over its own security and alliances, and no country would have a veto over another’s sovereign choices. NATO will remain central to European collective defense, and this initiative would build upon existing institutions and commitments. We will continue consultations through the Shell Bay format with Ukraine, NATO, the EU, and our European partners, expanding engagement to ensure that all relevant perspectives and ideas are considered. We will also engage Russia as part of the peace negotiations. Next week in Abu Dhabi, the United States will convene the first trilateral peace discussions with Ukraine and Russia since February. All parties remain intensely engaged in preparing for substantive and productive negotiations. We will consult closely with our European partners on the sidelines and, if progress is made, aim to move toward a broader multilateral format with their direct participation. These ideas continue to evolve, and we welcome thoughtful input. Creating the right structure will be critical to ending this war and transitioning from conflict toward conditions that make possible a durable peace. This is difficult, detailed work. There are no easy solutions. But the opportunity to help end this devastating war and build a foundation for a more peaceful, secure, and prosperous Europe for generations to come makes it a challenge well worth undertaking. We will see how these discussions develop and continue working intensively with all parties toward President Trump’s objective of a comprehensive and lasting peace.
123
1,967
122
7,223
369,846
I said at the time - and repeatedly since - that things changed in November 2025, that the pace sped up, and that we were on a new trajectory. There's no denying now that was true. So, for the record: things changed again in September 2026. We are on a new trajectory.
92
113
20
1,226
58,762
First ever post from Sam McCandlish. Anthropic co-founder and Chief Architect, leads pretraining and RL. He says Anthropic is preparing to share a report on their internal acceleration. Bet on this being the announcement they have hit the 2x Responsible Scaling Policy threshold.
I'm Sam, one of Anthropic's co-founders, and I lead much of our safety work. I'm going to have more to share on that work over the coming months. I thought I'd start by sharing a few thoughts on the latest report from @METR_Evals. The speed of progress in model capabilities is one of the most important questions in AI safety, and METR’s work here provides the public with an independent, external view into how this is happening inside Anthropic. I’ll talk more about some of the details below, but first some context for those new to this work. How quickly AI progress is accelerating has many important safety implications, since it determines how quickly the risks from advanced models increase. But, because most research happens within the frontier labs, it’s difficult for external observers and policy makers to get a strong sense of what that rate is. We kicked off the process with METR that led to this report earlier in the year, in anticipation of progress continuing to accelerate. For this report, we gave METR access to a range of sensitive internal data, including benchmark scores for unreleased models, estimates of how models are progressing, and interviews with our researchers. They had complete independence to form their own conclusions. Many of their conclusions match my own. AI is speeding up our research, but it’s hard to say exactly how much. While they conclude AI can't yet make the most important contributions to model training on its own, they think the pace of development could, by default, accelerate further. This work began well before and is separate from the embedded third-party evaluators that are part of our commitment to pacing the frontier, but it’s a good illustration of why we think evaluators are a good idea. The more access reviewers have, the clearer a picture they can form of the situation and the better-positioned they are to highlight situations like this one. We redacted a lot, and I accept that this makes it harder for others to check METR’s work. But we determined that sharing exactly what is driving the speedup isn’t the right thing to do, both because it’s commercially sensitive and because it risks handing out a roadmap for others to go faster. That would make the risks METR is worried about worse. That’s a tradeoff. I think we made the right call, though we’re continuing to assess the pros and cons here. In the coming weeks, we plan to share a report with our own analysis of the speed of AI progress within Anthropic, including where we stand against our Responsible Scaling Policy, plus new steps we’re taking to strengthen our safety and security practices as AI keeps improving quickly.
50
151
6
3,178
197,262